Chamanara's personal blog

تغذيه ماهي

2006-02-02T02:39:00.000+03:30

Understanding Fish Nutrition, Feeds, and Feeding

Authors: Steven Craig, Assistant Professor, Virginia-Maryland Regional College of Veterinary Medicine; and L. A. Helfrich, Extension Specialist and Professor, Department of Fisheries and Wildlife Sciences; Virginia Tech Publication Number 420-256, Posted April 2002

Introduction
Good nutrition in animal production systems is essential to economically produce a healthy, high quality product. In fish farming, nutrition is critical because feed represents 40-50% of the production costs. Fish nutrition has advanced dramatically in recent years with the development of new, balanced commercial diets that promote optimal fish growth and health. The development of new species-specific diet formulations supports the aquaculture (fish farming) industry as it expands to satisfy increasing demand for affordable, safe, and high-quality fish and seafood products.

Prepared (artificial) Diets
Prepared or artificial diets may be either complete or supplemental. Complete diets supply all the ingredients (protein, carbohydrates, fats, vitamins, and minerals) necessary for the optimal growth and health of the fish. Most fish farmers use complete diets, those containing all the required protein (18-50%), lipid (10-25%), carbohydrate (15-20%), ash (< 8.5%), phosphorus (< 1.5%), water (< 10%), and trace amounts of vitamins, and minerals. When fish are reared in high density indoor systems or confined in cages and cannot forage freely on natural feeds, they must be provided a complete diet.
In contrast, supplemental (incomplete, partial) diets are intended only to help support the natural food (insects, algae, small fish) normally available to fish in ponds or outdoor raceways. Supplemental diets do not contain a full complement of vitamins or minerals, but are used to help fortify the naturally available diet with extra protein, carbohydrate and/or lipid.
Fish, especially when reared in high densities, require a high-quality, nutritionally complete, balanced diet to grow rapidly and remain healthy.

Protein
Because protein is the most expensive part of fish feed, it is important to accurately determine the protein requirements for each species and size of cultured fish. Proteins are formed by linkages of individual amino acids. Although over 200 amino acids occur in nature, only about 20 amino acids are common. Of these, 10 are essential (indispensable) amino acids that cannot be synthesized by fish. The 10 essential amino acids that must be supplied by the diet are: methionine, arginine, threonine, tryptophan, histidine, isoleucine, lysine, leucine, valine and phenylalanine. Of these, lysine and methionine are often the first limiting amino acids. Fish feeds prepared with plant (soybean meal) protein typically are low in methionine; therefore, extra methionine must be added to soybean-meal based diets in order to promote optimal growth and health. It is important to know and match the protein requirements and the amino acid requirements of each fish species reared. Protein levels in aquaculture feeds generally average 18-20% for marine shrimp, 28-32% for catfish, 32-38% for tilapia, 38-42% for hybrid striped bass. Protein requirements usually are lower for herbivorous fish (plant eating) and omnivorous fish (plant-animal eaters) than they are for carnivorous (flesh-eating) fish, and are higher for fish reared in high density (recirculating aquaculture) than low density (pond aquaculture) systems. Protein requirements generally are higher for smaller fish. As fish grow larger, their
protein requirements usually decrease. Protein requirements also vary with rearing environment, water temperature and water quality, as well as the genetic composition and feeding rates of the fish. Protein is used for fish growth if adequate levels of fats and carbohydrates are present in the diet. If not, protein may be used for energy and life support rather than growth.
Proteins are composed of carbon (50%), nitrogen (16%), oxygen (21.5%), and hydrogen (6.5%). Fish are capable of using a high protein diet, but as much as 65% of the protein may be lost to the environment. Most nitrogen is excreted as ammonia (NH3) by the gills of fish, and only 10% is lost as solid wastes. Accelerated eutrophication (nutrient enrichment) of surface waters due to excess nitrogen from fish farm effluents is a major water quality concern of fish farmers. Effective feeding and waste management practices are essential to protect downstream water quality.

Lipids (fats)
Lipids (fats) are high-energy nutrients that can be utilized to partially spare (substitute for) protein in aquaculture feeds. Lipids supply about twice the energy as proteins and carbohydrates. Lipids typically comprise about 15% of fish diets, supply essential fatty acids (EFA) and serve as transporters for fat-soluble vitamins. A recent trend in fish feeds is to use higher levels of lipids in the diet. Although increasing dietary lipids can help reduce the high costs of diets by partially sparing protein in the feed, problems such as excessive fat deposition in the liver can decrease the health and market quality of fish.
Simple lipids include fatty acids and triacylglycerols. Fish typically require fatty acids of the omega 3 and 6 (n-3 and n-6) families. Fatty acids can be: a) saturated fatty acids (SFA, no double bonds), b) polyunsaturated fatty acids (PUFA, >2 double bonds), or c) highly unsaturated fatty acids (HUFA; > 4 double bonds). Marine fish oils are naturally high (>30%) in omega 3 HUFA, and are excellent sources of lipids for the manufacture of fish diets. Lipids from these marine oils also can have beneficial effects on human cardiovascular health. Marine fish typically require n-3 HUFA for optimal growth and health, usually in quantities ranging from 0.5-2.0% of dry diet. The two major EFA of this group are eicosapentaenoic acid (EPA: 20:5n-3) and docosahexaenoic acid (DHA:22:6n-3). Freshwater fish do not require the long chain HUFA, but often require an 18 carbon n-3 fatty acid, linolenic acid (18:3-n-3), in quantities ranging from 0.5 to 1.5% of dry diet. This fatty acid cannot be produced by freshwater fish and must be supplied in the diet. Many freshwater fish can take this fatty acid, and through enzyme systems elongate (add carbon atoms) to the hydrocarbon chain, and then further desaturate (add double bonds) to this longer hydrocarbon chain. Through these enzyme systems, freshwater fish can
manufacture the longer chain n-3 HUFA, EPA and DHA, which are necessary for other metabolic functions and as cellular membrane components. Marine fish typically do not possess these elongation and desaturation enzyme systems, and require long chain n-3 HUFA in their diets. Other fish species, such as tilapia, require fatty acids of the n-6 family, while still others, such as carp or eels, require a combination of n-3 and n-6 fatty acids Carbohydrates Carbohydrates (starches and sugars) are the most economical and inexpensive sources of energy for fish diets. Although not essential, carbohydrates are included in aquaculture diets to reduce feed costs and for their binding activity during feed manufacturing. Dietary starches are useful in the extrusion manufacture of floating feeds. Cooking starch during the extrusion process makes it more biologically available to fish. In fish, carbohydrates are stored as glycogen that can be mobilized to satisfy energy demands. They are a major energy source for mammals, but are not used efficiently by fish. For example, mammals can extract about 4 kcal of energy from 1 gram of carbohydrate, whereas fish can only extract about 1.6 kcal from the same amount of carbohydrate. Up to about 20% of dietary carbohydrates can be used by fish.

Vitamins
Vitamins are organic compounds necessary in the diet for normal fish growth and health. They often are not synthesized by fish, and must be supplied in the diet. The two groups of vitamins are water-soluble and fat-soluble. Water-soluble vitamins include: the B vitamins, choline, inositol, folic acid, pantothenic acid , biotin and ascorbic acid (vitamin C). Of these, vitamin C probably is the most important because it is a powerful antioxidant and helps the immune system in fish. The fat-soluble vitamins include A vitamins, retinols (responsible for vision); the D vitamins, cholecaciferols (bone integrity); E vitamins, the tocopherols (antioxidants); and K vitamins such as menadione (blood clotting, skin integrity). Of these, vitamin E receives the most attention for its important role as an antioxidant. Deficiency of each vitamin has certain specific symptoms, but reduced growth is the most common symptom of any vitamin deficiency. Scoliosis (bent backbone symptom) and dark coloration may result from deficiencies of ascorbic acid and folic acid vitamins, respectively.

Minerals
Minerals are inorganic elements necessary in the diet for normal body functions. They can be divided into two groups (macro-minerals and micro-minerals) based on the quantity required in the diet and the amount present in fish. Common macro-minerals are sodium, chloride, potassium and phosphorous. These minerals regulate osmotic balance and aid in bone formation and integrity. Micro-minerals (trace minerals) are required in small amounts as components in enzyme and hormone systems. Common trace minerals are copper, chromium, iodine, zinc and
selenium. Fish can absorb many minerals directly from the water through their gills and skin, allowing them to compensate to some extent for mineral deficiencies in their diet.

Energy and Protein
Dietary nutrients are essential for the construction of living tissues. They also are a source of stored energy for fish digestion, absorption, growth, reproduction and the other life processes. The nutritional value of a dietary ingredient is in part dependant on its ability to supply energy. Physiological fuel values are used to calculate and balance available energy values in prepared diets. They typically average 4, 4, and 9 kcal/g for protein, carbohydrate and lipid, respectively.
To create an optimum diet, the ratio of protein to energy must be determined separately for each fish species. Excess energy relative to protein content in the diet may result in high lipid deposition. Because fish feed to meet their energy requirements, diets with excessive energy levels may result in decreased feed intake and reduced weight gain. Similarly, a diet with inadequate energy content can result in reduced weight gain because the fish cannot eat enough feed to satisfy their energy requirements for growth. Properly formulated prepared feeds have a well-balanced energy to protein ratio.

Feed Types
Commercial fish diets are manufactured as either extruded (floating or buoyant) or pressure-pelleted (sinking) feeds. Both floating or sinking feed can produce satisfactory growth, but some fish species prefer floating, others sinking. Shrimp, for example, will not accept a floating feed, but most fish species can be trained to accept a floating pellet. Extruded feeds are more expensive due to the higher manufacturing costs. Usually, it is advantageous to feed a floating (extruded) feed, because the farmer can directly observe the feeding intensity of his fish and adjust feeding rates accordingly. Determining whether feeding rates are too low or too high is important in maximizing fish growth and feed use efficiency.
Feed is available in a variety of sizes ranging from fine crumbles for small fish to large (1/2 inch or larger) pellets. The pellet size should be approximately 20-30% of the size of the fish species mouth gape. Feeding too small a pellet results in inefficient feeding because more energy is used in finding and eating more pellets. Conversely, pellets that are too large will depress feeding and, in the extreme, cause choking. Select the largest sized feed the fish will actively eat.

Feeding Rate, Frequency, and Timing

Feeding rates and frequencies are in part a function of fish size. Small larval fish and fry need to be fed a high protein diet frequently and usually in excess. Small fish have a high energy demand and must eat nearly continuously and be fed almost hourly. Feeding small fish in excess is not as much of a problem as overfeeding larger fish because small fish require only a small amount of feed relative to the volume of water in the culture system.
As fish grow, feeding rates and frequencies should be lowered, and protein content reduced. However, rather than switching to a lower protein diet, feeding less allows the grower to use the same feed (protein level) throughout the grow-out period, thereby simplifying feed inventory and storage.
Feeding fish is labor-intensive and expensive. Feeding frequency is dependent on labor availability, farm size, and the fish species and sizes grown. Large catfish farms with many ponds usually feed only once per day because of time and labor limitations, while smaller farms may feed twice per day. Generally, growth and feed conversion increase with feeding frequency. In indoor, intensive fish culture systems, fish may be fed as many as 5 times per day in order to maximize growth at optimum temperatures. Many factors affect the feeding rates of fish. These include time of day, season, water temperature, dissolved oxygen levels, and other water quality variables. For example, feeding fish grown in ponds early in the morning when the lowest dissolved oxygen levels occur is not advisable. In contrast, in recirculating aquaculture systems where oxygen is continuously supplied, fish can be fed at nearly any time. During the winter and at low water temperatures, feeding rates of warmwater fishes in ponds decline and feeding rates should decrease proportionally.

Feed acceptability, palatability and digestibility vary with the ingredients and feed quality. Fish farmers pay careful attention to feeding activity in order to help determine feed acceptance, calculate feed conversion ratios and feed efficiencies, monitor feed costs, and track feed demand throughout the year. Published feeding rate tables are available for most commonly cultured fish species.
Farmers can calculate optimum feeding rates based on the average size in length or weight and the number of fish in the tank, raceway, or pond (see Hinshaw 1999, and Robinson et al. 1998). Farmed fish typically are fed 1-4% of their body weight per day.

Automatic Feeders
Fish can be fed by hand, by automatic feeders, and by demand feeders. Many fish farmers like to hand feed their fish each day to assure that the fish are healthy, feeding vigorously, and exhibiting no problems. Large catfish farms often drive feed trucks with compressed air blowers to distribute (toss) feed uniformly throughout the pond. There are a variety of automatic (timed) feeders ranging in design from belt feeders that work on wind-up springs, to electric vibrating feeders, to timed feeders that can be programmed to feed hourly and for extended periods. Demand feeders do not require electricity or batteries. They usually are suspended above fish tanks and raceways and work by allowing the fish to trigger feed release by striking a moving rod that extends into the water. Whenever a fish strikes the trigger, a small amount of feed is released into the tank. Automatic and demand feeders save time, labor and money, but at the expense of the vigilance that comes with hand feeding. Some growers use night lights and bug zappers to attract and kill flying insects and bugs to provide a supplemental source of natural food for their fish.

Feed Conversion and Efficiency Calculations:
Because feed is expensive, feed conversion ratio (FCR) or feed efficiency (FE) are important calculations for the grower. They can be used to determine if feed is being used as efficiently as possible.
FCR is calculated as the weight of the feed fed to the fish divided by the weight of fish growth. For example, if fish are fed 10 pounds of feed and then exhibit a 5 pound weight gain, the FCR is 10/ 5 = 2.0. FCRs of 1.5-2.0 are considered „good¾ growth for most species.
FE is simply the reciprocal of FCRs (1/FCR). In the example above, the FE is 5/10 = 50%. Or if fish are fed 12 pounds of feed and exhibit a 4 pound weight gain, the FE = 4/12 = 30%. FEs greater than 50% are considered „good¾ growth.
Fish are not completely efficient (FEs of 100 %, FCRs of 1.0). When fed 5 pounds of feed, fish cannot exhibit 5 pounds of growth because they must use some of the energy in feed for metabolic heat, digestive processing, respiration, nerve impulses, salt balance, swimming, and other living activities. Feed conversion ratios will vary among species, sizes and activity levels of fish, environmental parameters and the culture system used.

Feed Care and Storage
Commercial fish feed is usually purchased by large farms as bulk feed in truckloads and stored in outside bins. Smaller farms often buy prepared feed in 50-pound bags. Bag feed should be kept out of direct sunlight and as cool as possible. Vitamins, proteins, and lipids are especially heat sensitive, and can be readily denatured by high storage temperatures. High moisture stimulates mold growth and feed decomposition. Avoid unnecessary handling and damage to the feed bags which may break the pellets and create „fines¾ which may not be consumed by fish.
Feed should not be stored longer than 90 to 100 days, and should be inventoried regularly. Bags should not be stacked higher than 10 at a time. Older feed should be used first, and all feed should be regularly inspected for mold prior to feeding. All moldy feed should be discarded immediately. Mice, rats, roaches and other pests should be strictly controlled in the feed storage area, because they consume and contaminate feed and transmit diseases.

Medicated Feeds
When fish reduce or stop feeding, it is a signal to look for problems. Off-feed behavior is the first signal of trouble such as disease or water quality deterioration in the fish growing system. Relatively few therapeutic drugs are approved for fish by FDA (see Helfrich and Smith 2001), but some medicated feeds for sick fish are available. Although using medicated feeds is one of the easiest ways to treat fish, they must be used early and quickly because sick fish frequently will stop feeding.

Managing Fish Wastes
The most important rule in fish nutrition is to avoid overfeeding. Overfeeding is a waste of expensive feed. It also results in water pollution, low dissolved oxygen levels, increased biological oxygen demand, and increased bacterial loads. Usually, fish should be fed only the amount of feed that they can consume quickly (less than 25 minutes). Many growers use floating (extruded) feeds in order to observe feeding activity and to help judge if more or less feed should be fed.
Even with careful management, some feed ends up as waste. For example, out of 100 units of feed fed to fish, typically about 10 units of feed are uneaten (wasted) and 10 units of solid and 30 units of liquid waste (50% total wastes) are produced by fish. Of the remaining feed, about 25% is used for growth and another 25% is used for metabolism (heat energy for life processes). These numbers may vary greatly with species, sizes, activity, water temperature, and other environmental conditions.

Useful References
Food Intake in Fish. 2001. Houlihan, D., Bouiard, T. and Jobling, M., eds. Iowa State
University Press. Blackwell Science Ltd. 418 pp.
Fish Kills: Their Causes and Prevention. 2001. Helfrich L., and S. Smith. Viginia
Cooperative Extension Service Publication 420- 252. Website:
http://www.ext.vt.edu/pubs/fisheries/420-252/420-252.html
Feeding Catfish in Commercial Ponds. 1998. E. Robinson, M. Li, and M. Brunson.
Southern Regional Aquaculture Center, Fact Sheet # 181. Web Site:
http://www.msstate.edu/dept/srac/fslist.htm
Nutrient Requirements of Fish. 1993. Committee on Animal Nutrition. National Research
Council. National Academy Press. Washington D.C. 114 pp.
Nutrition and Feeding of Fish. 1989. Tom Lovell. Van Nostrand Reinhold, New York.
260 pp.
Principles of Warmwater Aquaculture. 1979. Robert R. Stickney. John Wiley and Sons,
New York. 375 pp.
Standard Methods for the Nutrition and Feeding of Farmed Fish and Shrimp. 1990.
Albert G.J. Tacon. Volume 1: The Essential Nutrients. Volume 2:Nutrient Sources and
Composition. Volume 3: Feeding Methods. Argent Laboratories Press. Redmond, WA.
Trout Production: Feeds and Feeding Methods. 1999. Southern Regional Aquaculture
Center, Fact Sheet # 223. Web Site: http://www.msstate.edu/dept/srac/fslist.htm

يه مقاله خوب در مورد بي تقارني گنادها در قزل آلاي رنگين كمان - جديد

2006-02-02T02:03:00.000+03:30

Asymmetry in sexual development of gonads in intersex rainbow trout

E. QUILLET*†, L. LABBE‡ AND I. QUEAU‡
*Unite´ ‘Ge´ne´tique des Poissons’, Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas cedex, France and ‡Salmoniculture Expe´rimentale Marine IFREMER-INRA, Le Drennec, BP 17, 29 450 Sizun, France (Received 23 June 2003, Accepted 27 January 2004)
Large-scale sampling of spontaneous rainbow trout Oncorhynchus mykiss intersexes indicated a strong asymmetry of gonad differentiation in XX females; the right gonad was more sensitive to the mutation-induced masculinization than the left one.

Key words: gonad differentiation; intersex; rainbow trout; sex determination.

Salmonids are gonochoristic species. Intersexuality is very infrequently observed under natural or normal breeding conditions. Less than 20 intersex specimens have been described in salmonids (Stewart, 1891; de Beer, 1924; Crawford, 1927; Turner, 1946; Gibbs, 1956; Benson, 1958; Uzmann & Hesselholt, 1958; Ross et al., 1963; Hitron & Bonham, 1977; Hikita & Hashimoto, 1978; Fraser, 1997; Barnes et al., 2001), most in Oncorhynchus species. In rainbow trout Oncorhynchus mykiss (Walbaum), as far as is known, only two intersex individuals were actually reported (Gibbs, 1956; Ross et al., 1963), maybe three if the report by de Beer (1924) is included. Very little is known about the physiological mechanisms of gonad differentiation of spontaneous intersex individuals recorded in gonochoristic species. In salmonids, as in many other species, differentiation of gonads can be changed from the primary sex to the other by early hormonal treatments. Entirely sexreversed specimens, as well as incompletely reversed ones can be experimentally induced (Baroiller et al., 1999). Some intersex individuals have become sexually mature, and produced functional gametes of both sexes (Chevassus et al., 1988). Usually, spontaneous intersex specimens in normally gonochoristic populations are considered as teratological. In the past few years, exposure during early life stages to both adverse thermal effluent and chemicals (endocrine disrupting factors) have thought to be responsible (Fraser, 1997; Luksiene et al., 2000; Vigano et al., 2001; Gercken & Sordyl, 2002; Lyndon, 2002). Genetic factors may also induce instability in gonad differentiation. In rainbow trout, a mutation in the genetic sex determination system, mal mutation, (Quillet et al., 2002) has recently been described, which is associated with the development of testicular tissue in the gonads of expected XX individuals. Some of the mal-carrying individuals may not be fully sex-reversed from female to male, and remain intersex. A similar observation was made in another gonochoristic species, the common carp Cyprinus carpio L. by Komen et al. (1992a, b).
Here the data collected on intersex rainbow trout during the genetic analysis of mal mutation (Quillet et al., 2002) were used for which the orientation of sexreversal (from female to male) was unambiguously known. This made it possible to test for asymmetry in gonadal development in response to the disturbance caused by the mal mutation.
Data were compiled from several experiments performed in order to elucidate the origin of an unexpected maleness that had been recorded within the mitogynogenetic progeny of a rainbow trout female. Briefly, the transmission of maleness was examined across four generations, using both conventional and meiotic and mitotic gynogenetic offspring. Results support the hypothesis that maleness is due to a mutation termed mal, that can override the usual XX mechanism of sex determination and induce the development of testicular tissue in the gonads of expected XX individuals (Quillet et al., 2002). The significant point for the present study is that maleness in experimental progeny was consistently associated with a small proportion of intersex individuals. Further observations in other similar experimental crosses (unpubl. data) confirmed that association between maleness and intersex condition is the rule in mal-carrying progeny. In total, a set of 6000 mal-carrying progeny, reared in two different experimental farms, were sexed by examination of gonads. All were expected to be XX individuals and originated from the following types of crosses (Quillet et al., 2002): between mal-carrying males and females, between mal-carrying females and hormonally sex-reversed XX control males, self-fertilized progeny of one intersex adult and meiotic and mitotic-gynogenetic progeny of mal-carrying females. Intersexes were recorded in all of the groups and the overall frequency was 5_4% (322 individuals, included in the present study).
Sex was determined primarily on juveniles (6–13 months old). Fish were sacrificed (lethal dose of anaesthetic), both gonads were dissected, slightly squashed and examined under a microscope. Because of the pedigree of the groups, all fish were expected to be genetic XX females and gonads were expected to differentiate into ovaries. Every gonad was checked for the presence of male areas, and accordingly scored. Gonads were scored as female when they looked like control ovaries (shape and density of oocytes), or when ovarian lamellae were clearly recognizable all along the gonad, despite the number of oocytes sometimes being reduced. Abnormal sex differentiation may be associated with partial sterility in some females, so the presence of recognizable lamellae (rather than oocytes) was preferred as an ultimate criterion to attribute female phenotypic sex. Gonads were classified as ovotestes when at least in one region of the gonad, female morphological characteristics had completely disappeared (no oocyte as well as no lamellae could be recognized), and were replaced by areas looking like control testes. Finally, gonads were classified as testes (fully sex reversed) when they looked like a normal testis along the entire length (homogeneous structure and smooth surface). Individuals were scored as 1148 E. QUILLET ET AL.
intersexes when at least one of the gonads was an ovotestis or when one ovary and one testis were simultaneously observed. In rainbow trout adults, apparent maleness in sexually mature fish, based on presence of male secondary sexual characteristics (e.g. the shape of jaw, colour of the skin and abdominal shape) may mask a condition of intersexuality in the fish. Thus, as in juveniles, sex was assessed only after internal examination of the gonads. Gonads were scored as ovaries, ovotestes and testes according to the same criteria as for juveniles. Special attention was paid to the presence of small residual ovarian regions in mature testes. In the strain used, males usually become mature 1 year earlier than females, making the mature male phenotype more prominent in young adults. The results are summarized in Table I. Detailed information for right and left gonads was available for 288 individuals. Within gonads, all cases of arrangement between testicular and ovarian regions were recorded. Testicular tissue could be restricted to very small area, or invade the whole gonad. It could be located in one single region (anterior, posterior or inside the ovarian tissue), or several testicular regions could appear at different places along the ovary. Intersexuality affected both gonads in a limited number of cases. Only 17% of intersexes (50 records) had two ovotestes. In other cases, the phenotypic sex was different according the gonad. In 198 fish, maleness was more pronounced in the right gonad than in the left one, v. only 40 fish with the opposite (significantly different from a 1 : 1 ratio; w2, d.f.¼1, P<0_001). color="#000000">mutation.

In total, 203 adults aged 2–4 years were sacrificed during maturation for direct examination of the gonads, among which 23 intersexes were recorded. All produced milt, and ovarian sections of the gonads were developing normally. Two fish simultaneously produced milt and mature ova at the time of scrutiny, and the gametes were successfully fertilized (self-fertilization or cross mating). Although the position of the gonads (right or left) was recorded in only 10 of these adults, masculinization was, as in juveniles, more pronounced in one gonad than in the other: one single adult had kept an intact right ovary while six had kept an intact left ovary. Published data indicate that all situations are possible with regard to gonad condition in salmonids: intersex specimens can have two ovotestes, one ovary (or testis) and one ovotestis or one testis and one ovary. Yet, observations are too few to give indication of possible asymmetry, if any. In the present study, intersex condition was interpreted as the result of partial masculinization of gonads of XX individuals by sex-modifying genetic factors. A strong lateral asymmetry in the sexual development of gonads was recorded, with a more pronounced sex reversal in the right gonad. Despite a limited number of observations of adults, it seems that the situation is stable throughout the life of intersexes. In normal juvenile rainbow trout, Mrsic (1923) noticed that the left ovary was longer and often ‘more developed’ than the right one. Ashby (1957) made the same observation (right ovary reduced in size) in brown trout Salmo trutta L. In juvenile mal-carrying female rainbow trout (with two normal ovaries) in the present study, the left ovary tended to be more developed than the right one (the left ovary was longer than right one in 21 females out of 43 examined, while the right one was longer in six cases only). All these observations support the hypothesis of a ‘stronger femaleness’ of the left gonad, and are consistent with the observations that the right gonad is more sensitive to the mutation-induced masculization. In contrast, however, following artificial masculinization with methyltestosterone in brown trout, Chevassus & Krieg (1992) observed the opposite, i.e. the right ovary was the most refractory to masculinization. Thus, asymmetry of gonads may be the rule generally in salmonids as in other vertebrate species (i.e. birds) but its expression will depend on the factors acting on it, and will generally lead to limited lateral differences. The authors sincerely thank the staff of experimental farms [INRA, Gournay-sur- Aronde (60) and SEMII, Sizun (29), France] for their efficient help with the work. Some of the experiments were jointly supported by IFREMER and INRA.

References

Ashby, K. R. (1957). The effect of steroid hormones on the brown trout (Salmo trutta L.) during the period of gonadal differentiation. Journal of Embryology and Experimental Morphology 5, 225–249. Barnes, M. E., Carreiro, J. M. & Cordes, R. J. (2001). Hermaphroditism observed in captive fall Chinnok salmon broodstock. North American Journal of Aquaculture 63, 262–264. Baroiller, J. F., Guiguen, T. & Fostier, A. (1999). Endocrine and environmental aspects of sex differentiation in fish. Cellular and Molecular Life Sciences 55, 910–931. de Beer, G. R. (1924). Note on a hermaphrodite trout. Anatomical Record 27, 61–62. Benson, N. G. (1958). Hermaphroditism in the cutthroat trout. Copeia 1958, 239–240. 1150 E. QUILLET ET AL. Chevassus, B. & Krieg, F. (1992). Effect of the concentration and duration of methyltestosterone treatment on masculinization rate in the brown trout (Salmo trutta). Aquatic Living Resources 5, 325–328. Chevassus, B., Devaux, A., Chourrout, D. & Jalabert, B. (1988). Production of YY rainbow trout males by self-fertilization of induced hermaphrodites. Journal of Heredity 79, 89–92. Crawford, D. R. (1927). Notice of hermaphroditism in silver salmon, Oncorhynchus kisutch. Copeia 1927, 34. Fraser, D. (1997). A hermaphrodite Arctic charr from Loch Rannoch, Scotland. Journal of Fish Biology 50, 1358–1359. doi: 10.1006/jfbi.1997.0392. Gercken, J. & Sordyl, H. (2002). Intersex in feral marine and freshwater fish from northeastern Germany. Marine Environmental Research 54, 651–655. doi: 10.1016/ S0141-1136(02)00156-3. Gibbs, E. D. (1956). A bisexual steelhead. California Fish and Game 42, 229–231. Hikita, T. & Hashimoto, S. (1978). A hermaphroditic chum salmon, Oncorhynchus keta, from the Chitose River with an example of its self-fertilization. Scientific Reports of the Hokkaido Salmon Hatchery 32, 61–64. Hitron, J. W. & Bonham, K. (1977). Hermaphroditism in a chum salmon (O. keta). Copeia 1977, 591–592. Komen, J., de Boer, P. & Richter, C. J. J. (1992a). Male sex reversal in gynogenetic XX females of common carp (Cyprinus carpio L.) by a recessive mutation in a sex-determining gene. Journal of Heredity 83, 431–434. Komen, J., Wiegertjes, G. F., Van Ginneken, V. J. T., Eding, E. H. & Richter, C. J. J. (1992b). Gynogenesis in common carp (Cyprinus carpio L.). III. The effect of inbreeding on gonadal development of heterozygous and homozygous offspring. Aquaculture 104, 51–66. Luksiene, D., Sandstro¨ m, O., Lounasheimo, L. & Andersson, J. (2000). The effects of thermal effluent exposure on the gametogenesis of female fish. Journal of Fish Biology 56, 37–50. doi: 10.1006/jfbi.1999.1139. Lyndon, A. (2002). Intersex fish found in U.K. estuaries. Marine Pollution Bulletin 44, 722. Mrsic, W. (1923). Die Spa¨ tbefruchtung und deren Einfluك auf Entwicklung und Geschlectsbildung, experimentell nachgepru¨ ft an der Regenbogenforelle. Archiv fu¨er mikroskopische Anatomie 98, 129–209. Quillet, E., Aubard, G. & Que´ au, I. (2002). Mutation in a sex-determining gene in rainbow trout: detection and genetic analysis. Journal of Heredity 93, 91–99. Ross, A. J., Yasutake, W. T. & White, G. R. (1963). Hermaphroditism in rainbow trout. Transactions of the American Fisheries Society 92, 313–314. Stewart, C. (1891). On a hermaphrodite trout, Salmo fario. Journal of the Linnean Society of London (Zoology) 24, 69–70. Turner, C. L. (1946). A case of hermaphroditism in the cutthroat trout. The Chicago Academy of Sciences, Natural History Miscellenea 1, 1–2. Uzmann, J. R. & Hesselholt, M. N. (1958). Teratological hermaphroditism in the chum salmon, Oncorhynchus keta (Walbaum). Progressive Fish-Culturist 20, 191–192. Vigano, L., Arillo, A., Massari, A. & Mandich, A. (2001). First observation of intersex cyprinids in the Po River (Italy). The Science of the Total Environment 269, 189–194. doi: 10.1016/S0048-9697(00)00821-4. INTERSEX IN RAINBOW TROUT 1151

# 2004 The Fisheries Society of the British Isles, Journal of Fish Biology 2004, 64, 1147–1151

Near infrared refectance spectroscopy in the prediction

2005-10-18T04:55:00.000+03:30

D. COZZOLINO1, I. MURRAY1 & J.R. SCAIFE2
1 Animal Biology, SAC Aberdeen, Scottish Agricultural College, Aberdeen, Scotland, UK; 2 Department of Agriculture,
MacRobert Building, University of Aberdeen, Aberdeen, Scotland, UK

Abstract

Near infrared refectance spectroscopy in the prediction
of chemical characteristics of minced raw ®sh
D. COZZOLINO1, I. MURRAY1 & J.R. SCAIFE2
1 Animal Biology, SAC Aberdeen, Scottish Agricultural College, Aberdeen, Scotland, UK; 2 Department of Agriculture,
MacRobert Building, University of Aberdeen, Aberdeen, Scotland, UK
Abstract
Near infrared re¯ectance spectroscopy (NIRS) was applied
to predict chemical composition in minced raw ®sh samples
used to make ®shmeal. The coecients of determination
(R2
calibration) and standard error in cross validation (SECV)
were 0.99 (3.86) and 0.96 (8.01) in g kg±1 for moisture and
oil, respectively. Total volatile nitrogen (TVN) gave R2
calibration
and SECV of 0.96 (3.51) in mg g±1. Temperature also was
predicted by NIRS, yielding R2
calibration: 0.98 and SECV:
1.07 °C. We conclude that NIRS can be used successfully to
assess the chemical composition and storage conditions in
minced raw ®sh used by the ®shmeal industry.
KEY WORDS : moisture, near infrared spectroscopy, oil, raw
®sh, temperature, total volatile nitrogen
Received 10 July 2000, accepted 12 December 2000
Correspondence: D. Cozzolino, National Institute for Agricultural Re-
search, INIA La Estanzuela. Ruta 50 km 11, CC 39173, Colonia, Uruguay.
E-mail: cozolino@inia.org.uy
Introduction
The quality of raw ®sh is highly variable in many properties
(moisture, oil, protein and volatile nitrogen from protein
breakdown). This variability arises from di.erent ®sh species,
®sh processing systems and seasonal variations. Recent
studies have shown that deteriorative changes in proteins
and lipids occur also during the storage of ®sh (Borquez &
Speek 1994; Raghunath et al. 1995), increasing variability.
Fish quality has traditionally been evaluated through sensory
assessments (Freeman & Hearnsberger 1994). However,
sensory assessments are subjective and a trained taste panel
is needed to carry out this evaluation. In the case of ®sh
waste from ®lleting operations, the use of a taste panel is not
appropriate. The search for a simple objective method to
assess chemical and physical characteristics has been receiv-
ing increasing attention during the past years (Zhang & Lee
1997). The lack of simple, reliable and nondestructive
methods for the determination of carcass composition in
®sh and ®sh by-products, has been one of the main obstacles
for the development of quality control in the ®sh industry.
Conventional methods involve time consuming, laborious
and costly procedures, including dissection and chemical
analysis. During recent years, new developments have
resulted in rapid methods relating multivariate physical
records of investigating samples to content of speci®c
chemical constituents. Consequently, the demand for tradi-
tional analysis using chemical reagents is reduced (Rye 1991).
Near infrared re¯ectance spectroscopy (NIRS) is a physical
and nondestructive technique. The NIR region of the
electromagnetic spectrum lies between the visible and infra-
red region (750±3000 nm), while the spectra appear as
smooth, but they consist of many overlapped bands.
The re¯ectance spectrum of a sample is the summation of
the spectra of its major chemical components (Deaville &
Flinn 2000). In raw ®sh, NIRS has successfully been applied
for determination of carcass composition in rainbow trout
(Gjerde & Martens 1987; Valdes et al. 1989), salmon (Lee
et al. 1992), freshwater ®sh (Mathias et al. 1987), ®sh tissue
using mid-infrared transmission (Darwish et al. 1989) and
near infrared transmittance (Solberg 1995). The total UK
production of ®shmeal in 1997 was 45 000 tonnes [mostly
from the trimming from food ®sh (UFP 1996; FIN 1998)]. A
large proportion of this production came from industrial ®sh
processing. Raw ®sh is ®lleted, skinned and trimmed to
produce ®llets, and these by-products are used to make the
®shmeal by the industry. To be able to control and optimize
the processing of the ®shmeal, it is important to measure and
analyse the chemical composition of the raw material. It
would be valuable to determine key parameters such as the
chemical composition (moisture, oil and protein) and other
parameters such as total volatile nitrogen (TVN) or salt
1
Aquaculture Nutrition 2002 8;1^6
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ó 2002 Blackwell Science Ltd
content related with both storage and conservation of the
material before entry into the ®shmeal factory. Several
chemical compounds have been found related to the change
of ®sh quality such as ATP degradation products (e.g.
hypoxanthine, trimethylamine, total volatile base, free fatty
acids). The methods for determination of those compounds
generally are colorimetric or chromatographic measure-
ments. These methods also required a lot of sample prepar-
ation and involve many chemical manipulations (Zhang &
Lee 1997). The objectives of this paper were (1) to report the
main absorption bands in the near infrared region used to
assess the chemical composition of minced raw ®sh and (2) to
predict moisture, oil, TVN and temperature in the samples
for quality control in the ®shmeal industry.
Materials and methods
Samples
One hundred and ®ve (n . 105) minced raw ®sh samples
from an industrial manufacturing plant (UFP, Tullos,
Aberdeen, UK) were collected from October 1996 to August
1997. Minced raw ®sh samples came from di.erent ®sh
species such as mackerel (n . 15) (Scomber scombrus),
herring (n . 25) (Clupea harengus), salmon (n . 15) (Salmo
salar), bluewhiting (n . 20) (Micromesistius poutassau) and
other ®sh species (n . 30). These samples were mainly
produced as a by-product of ®lleting ®sh for human
consumption, except bluewhiting. Whole bluewhiting sam-
ples were eviscerated and skinless, after homogenization. The
samples were homogenized fresh in a food blender, (Moul-
inex, France) and scanned fresh.
Chemical analysis
Moisture content was determinated by oven drying the
sample at 105 °C for 4 h (AOAC 1990), oil was extracted by
Soxhlet (Det, Gras, Selecta, Spain) using petroleum ether
(BP 40±60°) (AOAC 1990). The TVN was measured by
alkaline distillation and titration using 0.1 N NaOH until the
indicator turns from purple to green (AOAC 1990). Tem-
perature was measured in the fresh material using a digital
thermocouple.
NIR analysis
Aliquots (25±50 g) of minced raw ®sh were scanned fresh
from 1100 to 2500 nm in re¯ectance mode at 4 nm intervals
in a scanning monochromator NIRS 5000 (NIRSystems,
Silver Spring, MD, USA).
Computer operation and data was manipulated using ISI
version 3.1 software (InfraSoft International, Port Matilda,
PA, USA). The samples were placed in a large rectangular
quartz cup (Part Number IH-0314). Two pairs of lead
sulphide detectors collected the re¯ectance spectra. The
absorbance spectrum was recorded as log (1/R; R: re¯ect-
ance) for each minced raw ®sh sample. Re¯ected energy
readings were referenced to corresponding readings from a
ceramic disk. A reference scan was collected and stored to
computer memory before each sample was scanned. The
spectrum of each sample was the average of 32 successive
scans. Prediction models were developed using modi®ed
partial least squares regression (MPLS) (Shenk & Wester-
haus 1993) with cross validation and scatter correction by
standard normal variate (SNV) and detrend (Barnes et al.
1989). Because NIR spectra are a.ected by particle size and
light scatter (re¯ectance) and path-length variation (trans-
mission), pretreatment of the spectral data improve calibra-
tion accuracy (Deaville & Flinn 2000). Application of SNV
and detrend transformation to the spectral data results in
spectra which have reduced amounts of variation because of
physical e.ects (Sanderson et al. 1997). Cross validation was
used to avoid over®tting of the equations. Cross validation
estimates the prediction error by splitting the calibration
samples into groups (four in this study). One group was
reserved for validation and the remaining groups were used
for calibration. The process was repeated until all groups
have been used for validation at once (Shenk & Westerhaus
1993). After cross validation, the calibration is performed on
all samples using the number of factors that gave the
minimum standard error in cross validation (SECV). The
math treatment applied was 1, 4, 4, 1, where the ®rst number
is the order of the derivative (1 is ®rst derivative of log 1/R),
the second number is the gap in nm over which the derivative
is calculated, the third number is the number of nm used in
the ®rst smoothing and the fourth number refers to the
number of nm over which the second smoothing is applied.
Calibration statistics calculated include the standard error
of calibration (SEC), the coecient of multidetermination in
calibration (R2
calibration), the SECV and the coecient of
determination in cross validation (R2
validation) (Shenk &
Westerhaus 1993). The optimum calibrations were selected
on the basis of minimising the SECV. The SEC and SECV
were calculated as follows:
SEC and SECV . ..Y ÿ Yi.2=.n ÿ t ÿ 1..1=2
D. Cozzolino et al.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ó 2002 Blackwell Science Ltd Aquaculture Nutrition8;1^6
2
where SEC and SECV are standard error of calibration and
standard error in cross-validation, respectively; Y and Yi are
the predicted and observed values for sample i (TVN, oil;
moisture); n is the number of samples used to build the
calibration models; t is the number of partial least squares
(PLS) factors in the model (Shenk & Westerhaus 1993).
Principal components (PC) were computed on spectra, in
order to rede®ne the optical properties of ®sh samples. The
Mahalanobis distance of each spectrum with respect to the
average spectra was calculated. In this way a structuring of
samples according to spectral features is possible and each
sample can be graphically placed in a three-dimensional plot
de®ned by any three PC scores (Shenk & Westerhaus 1993).
The CENTER program ranks spectra in a ®le according to
their Mahalanobis distance (H statistics) from the average
spectra of the ®le using PC scores (NIRS 21995). In order to
visualize the relative spectral position of samples from
di.erent ®sh species, samples were graphically displayed by
means of the ®rst two (one and two) or second two (two and
three), with the SYMMETRY program of the same software
(ISI, 3.01).
Results and discussion
Spectra characterization
The mean spectrum of minced raw ®sh samples is shown in
Fig. 1. The mean spectrum showed absorption bands at
1200 nm related to carbon-hydrogen (CH) stretch second
overtone, at 1456 nm related to oxygen-hydrogen (OH)
stretch ®rst overtone, at 1730 nm related to CH stretch ®rst
overtone, at 1947 nm related to OH absorption because of
water content and at 2310 nm to CH combinations (Murray
1986). According to Osborne et al. (1993) second derivative
spectra have a trough corresponding to each peak in the
original spectra. Figure 2 showed the second derivative of the
mean spectrum. The second derivative of the mean spectrum
had absorption bands at 1166 nm related to CH stretch
second overtone, at 1332 nm related to CH overtones, at
1418 nm related to OH stretch ®rst overtone, at 1712 and
1780 nm related to CH stretch ®rst overtone, at 1940 nm to
water, at 2054 and 2178 nm related to protein and at
2304 nm to CH combinations (Murray 1986; Osborne et al.
1993). In the second derivative, the main SDs occurred in the
water region (1456 and 1947 nm). That means water is the
domain component, which a.ected the mean spectrum of
minced raw ®sh samples.
NIRS calibration statistics for raw minced fish
Table 1 showed the NIRS calibration statistics in minced raw
®sh for the chemical parameters analysed. The R2
calibration and
SECV were 0.99 (3.86), 0.96 (3.51), 0.99 (8.01) and 0.98 (1.07)
for moisture (M), TVN and oil in g kg±1, and temperature
(T) in °C, respectively. Gjerde & Martens (1987) reported
that NIR can be used to determine fat, moisture and protein
in freeze dried rainbow trout (Oncorhynchus mykiss).
They used a 19 ®lter NIR instrument, in the wavelength
range 1445±2350 nm and the reported prediction errors were
4.5, 3.5 and 5.0 g kg±1 for fat, moisture and protein,
respectively. Rasco et al. (1991) analysed cross sections of
frozen and thawed rainbow trout by NIRS re¯ectance
between 900 and 1800 nm and found prediction errors of
10, 3.7 and 18 g kg±1 for fat, moisture and protein, respect-
ively. Sollid & Solberg (1992) measured homogenized Atlan-
tic salmon paste of 23 mm thickness, by near infra red
transmittance (NIT) and found prediction errors of 7 g kg±1
for fat. Isaksson et al. (1995) analysed intact salmon ®llets
Figure 1 NIRS mean spectrum (line) and SD (dotted) of minced raw
®sh samples.
Figure 2 Second derivative of NIRS mean spectrum (line) and SD
(dotted line) of minced raw ®sh samples.
NIRS in the prediction of chemical characteristics in fish
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ó 2002 Blackwell Science Ltd Aquaculture Nutrition8;1^6
3
(Salmo salar) by NIRS re¯ectance using remote and on line
®bre optic probe, either on frozen or thawed ®llets. They
found SEs for prediction of 12.8 and 11.6 in g kg±1 for fat
and moisture, respectively, worked in the spectral range of
1100±2500 nm. The calibration statistics obtained in this
work on raw ®sh were consistent with those of the previous
authors. An acceptable calibration for TVN was observed.
This parameter is a measure of nitrogen degradation during
storage and protein breakdown in raw ®sh (Borquez & Speek
1994; Borquez et al. 1994; Raghunath et al. 1995) and is used
to determine the quality (freshness) of the raw ®sh, previous
to its entry into the ®shmeal factory. The results found in this
work showed that NIRS could be used to assess this
parameter under industrial conditions satisfactorily. Oil
content in the sample was also well predicted by NIRS.
Although good calibrations were obtained for oil, a high
number of outlier samples were observed (n . 50). The
presentation of the samples to the instrument played an
important role to obtain good NIRS calibration statistics.
The nonadequate device used to present the samples to the
monochromator could explain the number of sample outliers
observed on the calibration models for oil. This was
especially veri®ed for very liquid samples, like salmon.
The cuvette was placed vertically and this may cause oil
separation in the sample during scan collection. Secondly,
di.erent species were also used to perform the calibration
models; ®sh samples with di.erent oil content because of
processing and storage conditions, seasonal variations,
determining di.erent optical characteristics. Figure 3 plotted
the clusters related to di.erent species used on the calibra-
tion. Four clusters were observed, corresponding to salmon,
bluewhiting, herring and other ®sh species. Individual NIRS
calibrations for each raw ®sh species were not explored,
because it escaped the objectives of this paper. Sample
temperature was well measured by NIRS. Temperature of the
minced raw sample was another parameter used to reject
samples by the industry. The sensing of temperature with
NIRS may come as a by-product of other measurements.
Any time we collect a spectrum which includes water as one
of the components, we have the possibility to measure the
temperature of that water (Norris 1988). This is true because
the water absorption bands are sensitive to temperature
(Isaksson et al. 1989). The performance of the calibration for
moisture, TVN, oils and temperature in the minced raw ®sh
was evaluated by using the SECV/SD ratio. When the error
in calibration (SEC or SECV) exceeds one-third of the SD of
the population, regression models can be misleading (Murray
1993). As none of the values for this ratio are >0.33 the
calibration models for moisture (SECV/SD: 0.06); TVN
(SECV/SD: 0.10) and oil (SECV/SD: 0.18) were classi®ed as
good (see Table 2). Figures 4 & 5 show the NIRS predicted
Mean SD SEC R2
calibration SECV R2
validation n T
M 723.0 58.3 3.03 0.99 3.86 0.99 82 7
TVN 33.3 7.2 1.37 0.96 3.51 0.83 83 10
oil 111.1 43.1 1.11 0.99 8.01 0.96 50 5
Temp. 12.5 3.7 0.38 0.98 1.07 0.92 60 13
SD: Standard deviation, SEC: standard error of calibration, R2
calibration: coe¤cient of multidetermination in
calibration, SECV: standard error of cross validation, R2
validation: coe¤cient of determination in cross
validation, T: number of terms used to perform the calibration model, TVN: Total volatile nitrogen
(mg g)1), M: moisture (g kg)1), oil: (g kg)1),Temp: temperature (°C).
Table 1 NIRS calibration and cross
validation statistics for moisture (M),
oil, TVN and temperature in minced raw
®sh
Figure 3 Score plots of minced raw ®sh samples. BW: Bluewhiting;
HE: herring; Salmon; other species (mackerel).
Table 2 Performance of calibration for raw ®sh samples
R2
calibration SD SECV CV (%) SECV/SD
M 0.997 58.3 3.86 0.53 0.06
TVN 0.996 33.3 3.51 10.5 0.10
oil 0.991 43.1 8.01 7.2 0.18
Temp. 0.989 3.7 1.07 8.6 0.29
SD: Standard deviation, R2
calibration: coe¤cient of multidetermination in
calibration, SECV: standard error of cross validation, CV percentage:
(SECV/mean) ´ 100; TVN: Total volatile nitrogen (mg g)1), M: moisture
(g kg)1), oil: (g kg)1),Temp.: temperature (°C).
D. Cozzolino et al.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ó 2002 Blackwell Science Ltd Aquaculture Nutrition8;1^6
4
data vs. chemical data for moisture and TVN. Moisture on
bluewhiting samples tends to be overestimated by NIRS.
Sample characteristics could explain these results (whole
samples) such as surface moisture, freshness. Besides TVN
were underestimated on salmon samples (Fig. 5).
Conclusions
This work demonstrates that NIRS is a simple and easy
technique that can be used to successfully monitor the quality
of raw ®sh used to make ®shmeal. The operational meas-
urements described were adequately determined by NIRS
and improve speed of reporting and assist in decision-making
processes of management and process optimization. On the
other hand, NIRS does not completely replace all reference
analytical methods for oil quality and it is important to
maintain skill in reference analysis by lab sta.. The industry
would like to get more speci®c chemical information related
to the freshness of the sample, which still requires to be
checked periodically. Installation of NIR leads to release of
lab sta. time from routine quality control analysis which
allows more e.ort to be directed towards the establishment
of more sophisticated chemical and physical information on
the raw ®sh (origin, species, salt, adulterants) which may in
future be analysed by NIRS.
Acknowledgements
Ms A. Chree and K. Kreshaw (UFP, Tullos, Aberdeen, UK)
are thanked for providing the samples and technical assist-
ance.
References
AOAC (1990) Ocial Methods of Analysis of the Association of
Ocial Analytical Chemists. (Helrich, K. ed.), 15th edn, Published
by the Association of Ocial Analytical Chemists, Inc., Arlington,
Virginia, USA. 1298 pp.
Barnes, R.J., Dhanoa, M.S. & Lister, S.J. (1989) Standard normal
variate transformation and detrending of near infrared di.use
re¯ectance spectra. Appl. Spectrosc., 43, 772±777.
Borquez, R. & Speek, A.J. (1994) In¯uence of acetic acid preserva-
tion of chilean mackerel (Trachurus murphy) on ®sh meal samples.
J. Sci. Food Agric., 66, 187±192.
Borquez, R., Espinoza, M. & Ormeno, R. (1994) E.ects of storage
time and chemical preservatives on the total volatile basic nitrogen
content in chilean mackerel (Trachurus murphy) prior to ®sh meal
production. J. Sci. Food Agric., 66, 181±186.
Darwish, G.S., Van de Voort, F.R. & Smith, J.P. (1989) Proximate
analysis of ®sh tissue by mid-infrared transmission spectroscopy.
Can. J. Fish. Aquat. Sci., 46, 644±649.
Deaville, E.R. & Flinn, P.Ch. (2000) Near infrared (NIR) spectro-
scopy: an alternative approach for the estimation of forage quality
and voluntary intake. In: Chapter 15. Forage Evaluation in Ruminant
Nutrition (Givens, D.I., Owen, E., Axford, R.F.E. & Omed, H.M.
eds), pp. 301±321. CAB International, Wallingford, UK.
FIN (1998) Fishmeal Information Network. Fishmeal Update. The
Chamberlain Partnership. Grain and Feed Trade Association,
Peterborough, UK. 3 pp.
Freeman, D.W. & Hearnsberger, J.O. (1994) Rancidity in selected
sites of frozen cat®sh ®llets. J. Food Sci., 59, 60±63.
Figure 4 NIRS data vs. chemical data for moisture in minced raw
®sh samples.
Figure 5 NIRS data vs. chemical data for TVN in minced raw ®sh
samples.
NIRS in the prediction of chemical characteristics in fish
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ó 2002 Blackwell Science Ltd Aquaculture Nutrition8;1^6
5
Gjerde, B. & Martens, H. (1987) Predicting carcass composition of
rainbow trout by near infrared re¯ectance spectroscopy. J. Anim.
Breed. Genet., 104, 137±148.
Isaksson, T., Ellekjaer, M.H.R. & Hildrum, K.I. (1989) Determin-
ation of the previous maximum temperature of heat-treated
minced meat by near infrared re¯ectance spectroscopy. J. Sci.
Food Agric., 49, 385±387.
Isaksson, T., Togersen, G., Iversen, A. & Hildrum, K. (1995) Non
destructive determination of fat, moisture and protein in Salmon
®llets by use of near infrared di.use spectroscopy. J. Sci. Food
Agric., 69, 95±100.
Lee, M.H., Cavinato, A.G., Mayes, D.M. & Rasco, B.A. (1992) Non
invasive short wavelength near infrared spectroscopy method to
estimate the crude lipid content in the muscle of rainbow trout.
J. Agric. Food Chem., 40, 2176±2181.
Mathias, J.A., Williams, P.C. & Sobering, D.C. (1987) The deter-
mination of lipid and protein in freshwater ®sh using near infrared
re¯ectance spectroscopy. Aquaculture, 61, 303±311.
Murray, I. (1986) The NIR spectra of homologous series of organic
compounds. In: NIR/NIT Conference (Hollo, J., Ka.ka, K.J. &
Gonczy, J.L. eds), pp. 13±28. Akademiai Kiado, Budapest.
Murray, I. (1993) Forage analysis by near infrared spectroscopy. In:
Chapter 14. Sward Herbage Measurement Handbook (Davies, A.,
Baker, R.D., Grant, S.A. & Laidlaw, A.S. eds), pp. 285±312.
British Grassland Society, University of Reading, Reading, UK.
Norris, K.H. (1988) History, present state and future prospects for
near infrared spectroscopy. In: Chapter 1. Analytical Applications
of Spectroscopy (Creaser, C.S. & Davies, A.M.C. eds), Royal
Society of Chemistry, pp. 3±9.
Osborne, B.G., Fearn, T. & Hindle, P.H. (1993) Near Infrared
Spectroscopy in Food Analysis, 2nd edn. Longman Scienti®c and
Technical, New York.
Raghunath, M., Sankar, T.V., Ammu, K. & Devadasan, K. (1995)
Biochemical and nutritional changes in ®sh proteins during drying.
J. Sci. Food Agric., 67, 197±204.
Rasco, B.A., Miller, C.E. & King, T.L. (1991) Utilisation of NIR
spectroscopy to estimate the proximate composition of trout
muscle with minimal sample pre treatment. J. Agric. Food Chem.,
39, 67±72.
Rye, M. (1991) Prediction of carcass composition in Atlantic salmon
by computerized tomography. Aquaculture, 99, 35±48.
Sanderson, R., Lister, S.J., Dhanoa, M.S., Barnes, R.J. & Thomas,
C. (1997) Use of near infrared re¯ectance spectroscopy to predict
and compare the composition of carcass samples from young
steers. Anim. Sci., 65, 45±54.
Shenk, J.S. & Westerhaus, M.O. (1993) Analysis of Agriculture and
Food Products by Near Infrared Re¯ectance Spectroscopy. Mono-
graph. Infrasoft International, Port Matilda, PA, USA.
Solberg, C. (1995) Use of near infrared transmittance in quality
analysis of ®sh. In: Near Infrared Spectroscopy: the Future Waves
(Williams, P.C. & Davies, A.M.C. eds), NIR Publications,
Chichester, UK, pp. 591±596.
Sollid, H. & Solberg, C. (1992) Salmon fat content estimation by
near infrared transmission spectroscopy. J. Food Sci., 57, 792±793.
Valdes, E.V., Atkinson, J.L., Hilton, J.W. & Leeson, S. (1989) Near
infrared re¯ectance analysis of fat, protein and gross energy of
chicken and rainbow trout carcasses. Can. J. Anim. Sci., 69,
1087±1090.
UFP (United Fish Products) (1996) Statistics of Production. UFP,
Tullos, Aberdeen. Internal Report.
Zhang, H.-Z. & Lee, T.-Ch. (1997) Rapid near infrared spectros-
copy method determination of free fatty acids in ®sh and its
application in ®sh quality as

دوستان عزيز

2005-08-29T11:15:00.000+04:30

دوستان عزيز اگر انتقاد يا پيشنهادي در رابطه با اين وبلاگ داريد خواهش مي کنم فقط به ادرس زير ارسال نماييد.
v.chamanara@gmail.com

food and nutrition

2005-08-28T00:59:00.000+04:30

Aquaculture Magazine Sep/Oct 2001
Volume 27, Number 5
© 2001 Aquaculture Magazine
FISH, FEED & NUTRITION
Ronald W. Hardy
Professor and Director
Hagerman Fish Culture Experiment Station
University of Idaho
August 20, 2001
One Size Doesn’t Fit All
Dr. Randy MacMillan, President of
the National Aquaculture Association
and VP of Research and Environmental
Affairs for Clear Springs Foods, Inc., is
famous in Southern Idaho for two
statements applied to the process of
establishing limits on nutrient levels in
trout farm discharge water. First,
MacMillan suggested that regulating
agencies could create a model to
establish optimum operating parameters
to reduce effluent phosphorus levels in
trout farms and that he expected such a
model to follow the dictum of “garbage
in, gospel out.” Second, concerning the
proposal to develop universal guidelines
for Idaho trout farms and effluent
management, MacMillan was quoted in
the local press as saying that “One size
doesn’t fit all.” MacMillan’s pithy
remarks use only four and five words,
respectively, but they say a great deal
about the proposals. Entire university
fisheries’ programs now focus on
computer modeling for setting standards
for aquaculture and as a result a
generation of students in these programs,
are diligently working on
computer tans. Modeling is a valuable
fisheries management tool but unless
models are developed by people who
get real tans spending time in the field
or on vessels gaining hands-on experience
about the variability of data that
goes into models, the odds of these
computer models being used widely is
remote. Concerning the second
observation that one size doesn’t fit all,
MacMillan is dead on, especially in
connection with nutritional requirements
of fish, and fish feed formulations.
Feed Formulation For Closely
Related Farmed Species
People in this business are said to
fall into two general groups those that
want to consolidate requirements for
like species, and those that want to
separate nutrition standards for each
individual species. Aquaculture
nutritionists may fall into either group.
For example, some have proposed that
the dietary requirements of salmon and
trout are basically the same, and thus
feed formulations should be the same,
including semi-purified diets used in
research. This group is the ‘one size
fits all group’. Others believe that each
species of fish, even those that are
closely related, have slightly different
dietary requirements, at least for some
essential nutrients. I fall into the latter
group, especially since I began rearing
wild cutthroat trout in our laboratory.
Cutthroat trout (Oncorhynchus
clarkii) are closely related to rainbow
trout (O. mykiss), being in the same
genus and being native to waters west
of the continental divide in North
America (Behnke, 1992). The Lewis
and Clark expedition first recorded
cutthroat trout in 1805, hence the
species name clarkii. There are two
divergent subspecies, the westslope
cutthroat (O. c. lewisi) associated with
the Columbia River, and the
Yellowstone cutthroat (O. c. bouvieri),
found in the upper Snake River drainage.
Several years ago, the Hagerman
Station was asked to conduct a feeding
study with cutthroat trout using fish of
wild origin. We collected wild cutthroat
from the Blackfoot River in
eastern Idaho (a tributary of the Snake
River) with the assistance of the Idaho
Department of Fish and Game, collected
gametes from mature females
and males, and brought them back to
the Station for fertilization and incubation.
When the fish were at the first
feeding stage, we offered them trout
starter feed that we routinely use with
rainbow trout. The cutthroat fry readily
consumed the feed, and all went well
for several weeks. Then a small
number of fish in each tank began to die
each day. The number rose over the
next few days. By chance, Professor
Ron Roberts, the world authority on
fish diseases (and columnist for this
magazine) was visiting, and he advised
in diagnosing the problem. We checked
the fish for bacterial and viral diseases
and found nothing. We checked water
quality and everything was fine.
Finally, Roberts remarked that the fish
seemed to be exhibiting signs of
vitamin deficiency, specifically pyridoxine
deficiency. I confirmed this in a
heartbeat, and we immediately began
feeding fresh beef liver. The fish
improved noticeably in a day, and
completely recovered in three days. We
switched to feed produced by another
company and resumed using rainbow
trout feed. Within a week, the fish
again showed signs of deficiency. Back
to liver and recovery. We then made
our own feed, doubling the vitamin
premix, and resumed feeding this to the
cutthroat trout fry. This time the fish
flourished. After checking with state
and federal fish hatcheries in the region
with experience raising wild cutthroat
Aquaculture Magazine Sep/Oct 2001
Volume 27, Number 5
.
trout, we found that our experience was
common. All hatcheries reported
having significant problems using
rainbow trout starter feeds with wild
cutthroat, and they supplemented
commercial feed with liver until the fish
reached the fingerling stage, at which
time they seemed to do well on rainbow
trout feeds. My fish are now two years
old and average 2-3 lbs., and have been
fed diets containing elevated levels of
vitamins compared to the recommended
levels specified for rainbow trout diets.
What could be different between
rainbow and cutthroat trout fry that
might influence their dietary requirement
for vitamins? They are both
native to the lakes and streams of
western North America. Their diets are
similar, at least during the fry stages.
Yellowstone cutthroat are native to the
western slopes of the Rocky Mountains,
whereas westslope cutthroat are found
from the western slopes of the Rocky
Mountains to coastal areas, similar to
rainbow trout. Thus, in general, they
are found in areas lower in elevation.
Cutthroat and rainbow trout are closely
related, and can even hybridize to
produce rainbow-cutthroat crosses.
Could the dietary requirements for
certain vitamins of Yellowstone
cutthroat trout really be different from
that of westslope cutthroat or rainbow
trout? Could our observations be the
result of using fish of wild origin?
Cutthroat trout from hatchery
stocks may have already undergone a
form of genetic modification as
compared to fish that could not survive
when fed feeds formulated for rainbow
trout fry. The answers to these questions
await studies by fish nutritionists.
But clearly MacMillan’s statement that
one size doesn’t fit all applies to the
formulation of starter feeds for cutthroat
and rainbow trout.
This observation has important
implications in the context of restoration
efforts being undertaken by
agencies to rebuild stocks of cutthroat
trout and other species in areas where
their numbers have been reduced,
especially efforts to rebuild fish
populations using captive broodstock
techniques with wild fish as the founder
stocks.
New York Times And Salmon
Aquaculture In Scotland
The New York Times ran a story on
July 17 on salmon aquaculture and
salmon sport fishing in Scotland,
written by Alan Cowell, a reporter from
London. The focus of the story was
that salmon fishing in Scotland has
diminished greatly as a result of weak
stocks. Many people blame salmon
farming for the reduction in wild
salmon stock abundance. I spoke to
Mr. Cowell for this article about salmon
aquaculture and found him to be an
open-minded, bright man who was
interested in getting a range of perspectives
on the pros and cons of salmon
farming. One issue we discussed at
length was the suggestion that increased
production of salmon by the aquaculture
industry is resulting or will result
in higher rates of harvest of wild, forage
fish to make fish meal for salmon feeds.
This is a common allegation by nongovernment
organizations (NGOs) that
oppose fish farming. I explained to Mr.
Cowell that fish meal production has
not increased as a result of salmon
farming, at least on a global basis.
However, fish feed producers are
purchasing a larger proportion of annual
global production of fish meal than they
did a decade ago because they are
willing to pay more for fish meal than
other major users, namely poultry and
swine feed producers. I also said that
given the fact fish meal is a finite
resource, this trend of a higher proportion
of annual global fish meal being
used to make fish feeds can only go on
so long. I explained that people in the
aquaculture industry were well aware of © 2001 Aquaculture Magazine

fizalia

2005-08-27T23:36:00.000+04:30

hello
the fizalia blog is created